How vaccine safety is monitored before and after approval — VAERS, adverse events, and benefit vs. risk assessment.
The U.S. operates three parallel post-market safety systems that form a detection-to-verification pipeline. VAERS (Vaccine Adverse Event Reporting System) is a passive reporting system — anyone can submit a report, making it the broadest signal detector. The Vaccine Safety Datalink (VSD) is an active surveillance system covering approximately 12 million people through electronic health records, allowing controlled comparisons between vaccinated and unvaccinated groups. The FDA's BEST system (Biologics Effectiveness and Safety System) uses large-scale insurance claims data for rapid safety studies. When VAERS detects a potential signal, VSD and BEST conduct controlled studies to determine whether the signal represents a genuine increased risk or is consistent with background rates. For detailed information, see our Post-Market Surveillance page at /regulatory/post-market-surveillance.
VAERS is a passive surveillance database where anyone — healthcare providers, manufacturers, or the public — can submit a report of any medical event that occurred after vaccination, regardless of whether the vaccine caused it. A VAERS report is not a confirmed injury. VAERS is designed for signal detection, not causation determination. Its strength is scope: it can detect unexpected patterns across millions of doses. Its limitation is that raw VAERS counts cannot be used to calculate injury rates without knowing total doses administered and without controlling for background rates of the reported conditions. Signals detected through VAERS are investigated by active surveillance systems (VSD and BEST) to determine whether they are genuine.
Yes. The identification of myocarditis risk following mRNA COVID-19 vaccines followed the detection-to-verification pipeline exactly. Step 1: VAERS received an unusual number of myocarditis reports in young males following mRNA vaccination. Step 2: VSD and BEST conducted controlled studies comparing myocarditis rates in vaccinated vs. unvaccinated populations, confirming an elevated risk of approximately 1–4 cases per 100,000 doses in males aged 12–29 after the second dose. Step 3: Regulatory agencies updated vaccine guidance and labeling accordingly. Most cases were mild and resolved within days to weeks. See our Myocarditis page at /safety/myocarditis.
Yes. All vaccines go through Phase 1, 2, and 3 clinical trials before approval — a process that typically takes 10–15 years. Phase 3 trials alone enroll thousands to tens of thousands of participants specifically to detect safety signals. After approval, safety monitoring continues indefinitely through the VAERS, VSD, and BEST systems described above.
A side effect is an expected, known reaction to a vaccine — such as a sore arm or mild fever. This is part of the body's normal immune response, known as reactogenicity. An adverse event is any medical occurrence after vaccination, whether or not it is causally related to the vaccine. VAERS collects adverse event reports; not all reported events are caused by the vaccine. The distinction between expected reactogenicity and unexpected adverse events is central to how the safety monitoring system operates.
With inactivated, subunit, mRNA, and viral vector vaccines — no. These vaccines contain no live pathogen capable of causing infection. With live attenuated vaccines (MMR, varicella, nasal spray flu), there is a very rare theoretical risk. Vaccine-derived polio from the oral live vaccine has occurred in under-vaccinated populations, which is why the U.S. switched to the inactivated polio vaccine (IPV) in 2000.
No. This is one of the most thoroughly studied questions in vaccine science. The original 1998 Wakefield study claiming an MMR-autism link was retracted by The Lancet in 2010 following findings of research misconduct. Over 30 large epidemiological studies across multiple countries, involving millions of children, have found no association between any vaccine and autism. The U.S. Vaccine Court dismissed over 5,500 autism-related petitions after reviewing the scientific evidence. See our Autism & MMR Controversies page at /controversies/autism-mmr.
Thimerosal contains ethylmercury, which is metabolized and eliminated from the body rapidly — unlike methylmercury (found in fish), which accumulates. Thimerosal was phased out of routine childhood vaccines in the U.S. between 1999 and 2001 as a precautionary measure. Autism rates continued to rise after removal, providing further evidence against any causal link. Thimerosal remains in some multi-dose influenza vaccine formulations. For detailed information, see our Thimerosal page at /research/vaccines/ingredients/thimerosal.
Vaccine ingredients include antigens, adjuvants, preservatives, stabilizers, residual manufacturing materials, and diluents. All are present in quantities far below established safety thresholds. Aluminum adjuvants, for example, are present in microgram quantities — far less than the daily aluminum intake from food and water. Each ingredient has a documented safety profile based on decades of use and study. For detailed information, see our Vaccine Ingredients page at /research/vaccines/ingredients.
Yes. COVID-19 vaccines underwent the same Phase 1, 2, and 3 clinical trial process as all other vaccines. Phase 3 trials enrolled 30,000–44,000 participants per vaccine. The accelerated timeline was achieved through parallel development, pre-investment, and running trial phases concurrently — not by skipping any phase or shortening required follow-up periods. See our Clinical Trials page at /science/clinical-trials.
A small increased risk of myocarditis (inflammation of the heart muscle) has been confirmed following mRNA COVID-19 vaccination, primarily in males aged 12–29 after the second dose. The rate is approximately 1–4 cases per 100,000 doses in this group. Most cases are mild and resolve within days to weeks. The myocarditis risk from COVID-19 infection itself is substantially higher than from vaccination. This risk was identified through the VAERS→VSD detection-to-verification pipeline described above.